Skip to contentSkip to conversation

x86-refcount fix for v4.14-rc3

GIT PULL Kees Cook 2 participants 4 messages Sep 27, 2017

lore
Share & tools
lore

Kees Cook

author
#0 lore raw

Hi,

Please pull this x86-refcount fix for v4.14-rc3. This restores the x86 fast-refcount protection after fixing a corner case that was found late before the merge window opened. This was the only problem seen during the protection's life in linux-next, so getting this into v4.14 is highly desired.

Thanks!

-Kees

The following changes since commit e19b205be43d11bff638cad4487008c48d21c103:

Linux 4.14-rc2 (2017-09-24 16:38:56 -0700)

are available in the git repository at:

git://git.kernel.org/pub/scm/linux/kernel/git/kees/linux.git tags/x86-refcount-v4.14-rc3

for you to fetch changes up to 635fee8f6b725a99f511b8e0eca82c3214b9c678:

locking/refcounts, x86/asm: Enable CONFIG_ARCH_HAS_REFCOUNT (2017-09-27 03:05:13 -0700)

----------------------------------------------------------------
Fix x86 fast-refcount protection to avoid corner-case of being used a function living in the .text.unlikely section.

----------------------------------------------------------------
Kees Cook (2):
locking/refcounts, x86/asm: Use unique .text section for refcount exceptions locking/refcounts, x86/asm: Enable CONFIG_ARCH_HAS_REFCOUNT

 arch/x86/Kconfig                  | 2 +-
 arch/x86/include/asm/refcount.h   | 2 +-
 arch/x86/mm/extable.c             | 7 ++++++-
 include/asm-generic/vmlinux.lds.h | 1 +
 4 files changed, 9 insertions(+), 3 deletions(-)
Signature · 2 lines
Kees Cook
Pixel Security
#1 lore raw
Quoted 7 lines · Kees Cook

Hi,

Please pull this x86-refcount fix for v4.14-rc3. This restores the x86 fast-refcount protection after fixing a corner case that was found late before the merge window opened. This was the only problem seen during the protection's life in linux-next, so getting this into v4.14 is highly desired.

No, this is clearly v4.15 material.

Quoted 3 lines

Kees Cook (2):
locking/refcounts, x86/asm: Use unique .text section for refcount exceptions locking/refcounts, x86/asm: Enable CONFIG_ARCH_HAS_REFCOUNT

Do these commits differ from what you sent as patches on Sep 2?

Thanks,

Ingo

author
#2 lore raw
Quoted 5 lines · Ingo Molnar
Quoted 2 lines · Kees Cook

Hi,

Quoted 7 lines
Quoted 5 lines

Please pull this x86-refcount fix for v4.14-rc3. This restores the x86 fast-refcount protection after fixing a corner case that was found late before the merge window opened. This was the only problem seen during the protection's life in linux-next, so getting this into v4.14 is highly desired.

No, this is clearly v4.15 material.

Given the long bake time in linux-next (with only the one bug), and that the fix was done weeks ago, why the delay?

Quoted 5 lines
Quoted 3 lines

Kees Cook (2):
locking/refcounts, x86/asm: Use unique .text section for refcount exceptions locking/refcounts, x86/asm: Enable CONFIG_ARCH_HAS_REFCOUNT

Do these commits differ from what you sent as patches on Sep 2?

They are the same, but now that we've reached -rc2, there's no longer the fuzz against the linker script changes, so it seemed like a better way to send them your way.

-Kees

Signature · 2 lines
Kees Cook
Pixel Security
#3 lore raw
Quoted 12 lines · Kees Cook
Quoted 11 lines · Ingo Molnar
Quoted 7 lines · Kees Cook

Hi,

Please pull this x86-refcount fix for v4.14-rc3. This restores the x86 fast-refcount protection after fixing a corner case that was found late before the merge window opened. This was the only problem seen during the protection's life in linux-next, so getting this into v4.14 is highly desired.

Quoted 4 lines
Quoted 1 line

No, this is clearly v4.15 material.

Given the long bake time in linux-next (with only the one bug), and that the fix was done weeks ago, why the delay?

Because it's essentially enabling a new feature and the merge window is over. I'll apply them in 1-2 days.

Thanks,

Ingo